Tethered spinal cord: update of an analysis of published articles.

Two hundred and fifteen papers written in English between 1943-1994 addressing the issue of tethered spinal cord (TSC) were reviewed using established criteria for research design. Of 127 articles that were not reviews or case reports, 64% specified the diagnostic criteria used for TSC: only 6% discussed the representativeness of the sample. Observer bias was addressed in 6%: reliability in a mere 5%. Only 16% of the 43 articles dealing with clinical trials used a control group: in no case was it randomly chosen. Only 20% of the 25 articles addressing clinical course/prognosis utilized an inception cohort. A "gold standard" diagnostic test was used in 66% of the 44 studies dealing with the issue, but only 21% employed an independent blind comparison. Critical questions such as: What is the optimal surgery to prevent TSC? What is the optimal technique to repair TSC? Should surgery be performed on all asymptomatic children with TSC? Are changes in the appearance of the spinal cord associated with neurological deterioration? What is the role of TSC in scoliosis? cannot be answered because reliable data to answer them does not exist.